Original scientific paper
STROKE VOLUME AND PULSE PRESSURE VARIATION ARE GOOD PREDICTORS OF FLUID RESPONSIVENESS IN SEPSIS PATIENTS
ŽELJKO DRVAR
; Zagreb University Hospital Center, School of Medicine, University of Zagreb, Department of Anesthesiology, Resuscitation and Intensive Care, Zagreb, Croatia
MARIO PAVLEK
; Zagreb University Hospital Center, School of Medicine, University of Zagreb, Department of Anesthesiology, Resuscitation and Intensive Care, Zagreb, Croatia
VILIM DRVAR
; The City of Zagreb Institute of Emergency Medicine, Zagreb, Croatia
BORIS TOMAŠEVIĆ
; Zagreb University Hospital Center, School of Medicine, University of Zagreb, Department of Anesthesiology, Resuscitation and Intensive Care, Zagreb, Croatia
ROBERT BARONICA
; Zagreb University Hospital Center, School of Medicine, University of Zagreb, Department of Anesthesiology, Resuscitation and Intensive Care, Zagreb, Croatia
MLADEN PERIĆ
; Zagreb University Hospital Center, School of Medicine, University of Zagreb, Department of Anesthesiology, Resuscitation and Intensive Care, Zagreb, Croatia
Abstract
Stroke volume variation (SVV) and pulse pressure variation (PPV) are dynamic preload indicators. Specific interactions of the cardiovascular system and lungs under mechanical ventilation cause cyclic variations of SVV and PPV. Real time measurement of SVV and PPV by arterial pulse contour analysis is useful to predict volume responsiveness in septic patients. Results of a prospective, 2-year observational study conducted at Department of Anesthesiology, Resuscitation and Intensive Care, Zagreb University Hospital Center, are presented. Volume responders and non-responders were defined. Correlation between SVV, PPV, stroke volume index (SVI) and other hemodynamic data in septic patients was analyzed. The study was conducted from September 2009 to December 2011. Sepsis group included 46 patients (22 male, age 46±6, APACHE II score 26±5, and 24 female, age 41±6, APACHE II score 24±4) undergoing major abdominal surgery with clinically and laboratory confirmed sepsis, defined according to the international criteria. Exclusion criteria: patients with LVEF <45%, atrial fibrillation, aortic insufficiency, pulmonary edema, children, pregnant women, patients on lithium therapy, and patients who did not sign informed consent. Septic patients were divided into volume responders (VR) and volume non-responders (VNR). Responders were defined as patients with an increase in SVI of ≥15% after fluid loading. SVV, PPV and SVI were assessed by arterial pulse contour analysis using the LiDCOTM plus system continuously for 8 hours. Simultaneously, cardiac index (CI), mean arterial pressure (MAP), heart rate (HR), oxygen delivery (DO2), oxygen consumption (VO2) and central venous oxygen saturation (ScvO2) were assessed. Hemodynamic data were recorded before and after fluid administration of 500 mL of 6% hydroxyethyl starch over 30 min. All patients were sedated with midazolam (0.05-0.15 mg/kg/h). Analgesia was maintained with sufentanil (0.2-0.6 μg/kg). All patients were intubated and mechanically ventilated (IPPV; FiO2 0.4; TV 7 mL/kg; PEEP 5 cm H2O) in sinus cardiac rhythm. Circulatory unstable patients had vasoactive support and SOFA scores calculated. Ventilator settings and dosage of vasoactive drugs were all kept constant during the study. Data were compared using Student’s t-test. Correlation was estimated using Pearson’s coefficient. The level of statistical significance was set at P<0.05. Positive response to fluid loading was present in 26 (57.4%) patients. Baseline SVV correlated with baseline PPV (r=0.92, P<0.001). SVV and PPV were significantly higher in responders than in nonresponders. SVV: 14.4±3.3 vs. 7.1±3.1; P<0.001. PPV: 15.2±4.1 vs. 7.4±4.5; P<0.001. Other hemodynamic parameters measured were statistically different between the two groups. Only DO2 values showed no statistical significance between the responders and non-responders. There was no difference between the area under receiver operating characteristic curves of SVV (0.96; 95% confidence interval 0.859-0.996) and PPV (1.000; 95% confidence interval 0.93-1.000). Optimal threshold value for discrimination between VR and VNR was 10% for SVV (sensitivity 96.15%, specificity 100%) and 12% for PPV (sensitivity 100%, specificity 100%). In conclusion, SVV and PPV measured by LiDCOTM plus system are reliable predictors of fluid responsiveness in mechanically ventilated septic patients in sinus cardiac rhythm.
Keywords
stroke volume index; stroke volume variation; pulse pressure variation; fluid therapy; sepsis
Hrčak ID:
117908
URI
Publication date:
19.3.2014.
Visits: 2.382 *